The objectives of this project are to test two hypotheses. 1.AIDS patients who have weights loss without evidence of malabsorption, diarrhea, impaired intake or active infection are hypermetabolic. Metabolic processes which have been shown to be increased during hypermetabolism include protein turnover, acute phase protein synthesis, glucose and lipid substrate cycling and energy expenditure. We will determine whether these parameters are elevated in 12 AIDS patients. As comparisons, we will use 12 healthy normal controls and 12 HIV positive but otherwise asymptomatic patients. We will measure protein turnover and acute phase protein synthesis (fibrinogen) with 15N glycine, glucose substrate cycling with 6,6-D and 2-D labelled glucose, free fatty acid - triglyceride cycling with 1-13C palmitate and 5-D glycerol, and the resting energy expenditure with a metabolic cart. 2.A pilot study suggests that ambulatory weight stable AIDS patients without obvious evidence of malabsorption or active infection appear to exhibit a starvation response. We suggest that they have adapted to a lower rate of energy expenditure and balance in 12 such AIDS patients. The results will be compared against those found with 12 HIV positive but otherwise asymptomatic subjects and 12 normal controls,. Intake will be measured by dietary records over a five day period, energy expenditure by the Doubly Labelled Water method, balance from the difference between energy intake and expenditure. If hypothesis #2 is correct, the patients should be in energy balance, but their intake and energy expenditure less that either group of controls.